STEP 5: Plot the Veery observations by month

First thing first – let’s load your stored variables and import libraries.

%store -r

Try It: Import packages

In the imports cell, we’ve included some packages that you will need. Add imports for packages that will help you:

1. Make interactive maps with vector data

# Get month names
import calendar

# Libraries for Dynamic mapping
import cartopy.crs as ccrs
import panel as pn

# Get month names
import calendar

# Libraries for Dynamic mapping
import cartopy.crs as ccrs
import hvplot.pandas
import panel as pn

ERROR 1: PROJ: proj_create_from_database: Open of /usr/share/miniconda/envs/learning-portal/share/proj failed

Create a simplified GeoDataFrame for plotting

Plotting larger files can be time consuming. The code below will streamline plotting with hvplot by simplifying the geometry, projecting it to a Mercator projection that is compatible with geoviews, and cropping off areas in the Arctic.

Try It: Simplify ecoregion data

Download and save ecoregion boundaries from the EPA:

1. Simplify the ecoregions with .simplify(.05), and save it back to the geometry column.
2. Change the Coordinate Reference System (CRS) to Mercator with .to_crs(ccrs.Mercator())
3. Use the plotting code that is already in the cell to check that the plotting runs quickly (less than a minute) and looks the way you want, making sure to change gdf to YOUR GeoDataFrame name.

# Simplify the geometry to speed up processing

# Change the CRS to Mercator for mapping

# Check that the plot runs in a reasonable amount of time
gdf.hvplot(geo=True, crs=ccrs.Mercator())

# Simplify the geometry to speed up processing
ecoregions_gdf.geometry = ecoregions_gdf.simplify(
    .05, preserve_topology=False)

# Change the CRS to Mercator for mapping
ecoregions_gdf = ecoregions_gdf.to_crs(ccrs.Mercator())

# Check that the plot runs
ecoregions_gdf.hvplot(geo=True, crs=ccrs.Mercator())

Try It: Map migration over time

1. If applicable, replace any variable names with the names you defined previously.
2. Replace column_name_used_for_ecoregion_color and column_name_used_for_slider with the column names you wish to use.
3. Customize your plot with your choice of title, tile source, color map, and size.

Note

Your plot will probably still change months very slowly in your Jupyter notebook, because it calculates each month’s plot as needed. Open up the saved HTML file to see faster performance!

# Join the occurrences with the plotting GeoDataFrame
occurrence_gdf = ecoregions_gdf.join(occurrence_df)

# Get the plot bounds so they don't change with the slider
xmin, ymin, xmax, ymax = occurrence_gdf.total_bounds

# Plot occurrence by ecoregion and month
migration_plot = (
    occurrence_gdf
    .hvplot(
        c=column_name_used_for_shape_color,
        groupby=column_name_used_for_slider,
        # Use background tiles
        geo=True, crs=ccrs.Mercator(), tiles='CartoLight',
        title="Your Title Here",
        xlim=(xmin, xmax), ylim=(ymin, ymax),
        frame_height=600,
        widget_location='bottom'
    )
)

# Save the plot
migration_plot.save('migration.html', embed=True)

# Show the plot
migration_plot

# Join the occurrences with the plotting GeoDataFrame
occurrence_gdf = ecoregions_gdf.join(occurrence_df)

# Get the plot bounds so they don't change with the slider
xmin, ymin, xmax, ymax = occurrence_gdf.total_bounds

# Define the slider widget
slider = pn.widgets.DiscreteSlider(
    name='month', 
    options={calendar.month_name[i]: i for i in range(1, 13)}
)

# Plot occurrence by ecoregion and month
migration_plot = (
    occurrence_gdf
    .hvplot(
        c='norm_occurrences',
        groupby='month',
        # Use background tiles
        geo=True, crs=ccrs.Mercator(), tiles='CartoLight',
        title="Veery migration",
        xlim=(xmin, xmax), ylim=(ymin, ymax),
        frame_height=600,
        colorbar=False,
        widgets={'month': slider},
        widget_location='bottom'
    )
)

# Save the plot
migration_plot.save('migration.html', embed=True)

# Show the plot
migration_plot

  0%|          | 0/12 [00:00<?, ?it/s]  8%|▊         | 1/12 [00:00<00:01,  6.67it/s] 17%|█▋        | 2/12 [00:00<00:01,  6.03it/s] 25%|██▌       | 3/12 [00:00<00:03,  2.57it/s] 33%|███▎      | 4/12 [00:01<00:03,  2.03it/s] 42%|████▏     | 5/12 [00:02<00:03,  2.00it/s] 50%|█████     | 6/12 [00:02<00:03,  2.00it/s] 58%|█████▊    | 7/12 [00:03<00:02,  1.92it/s] 67%|██████▋   | 8/12 [00:03<00:02,  1.67it/s] 75%|███████▌  | 9/12 [00:04<00:01,  1.76it/s] 83%|████████▎ | 10/12 [00:04<00:00,  2.26it/s] 92%|█████████▏| 11/12 [00:04<00:00,  2.69it/s]100%|██████████| 12/12 [00:05<00:00,  3.15it/s]                                               

Looking for an Extra Challenge?: Fix the month labels

Notice that the month slider displays numbers instead of the month name. Use pn.widgets.DiscreteSlider() with the options= parameter set to give the months names. You might want to try asking ChatGPT how to do this, or look at the documentation for pn.widgets.DiscreteSlider(). This is pretty tricky!